32. GEOLOGIC HISTORY OF THE SOUTHEASTERN GULF OF MEXICO1
Wolfgang Schlager, School of Marine and Atmospheric Science, University of Miami, Miami, Florida Richard T. Buffler, David Angstadt,2 and Ronald Phair,3 Institute for Geophysics, University of Texas at Austin, Austin, Texas
ABSTRACT We have combined a seismic stratigraphic analysis of the southeastern Gulf of Mexico with DSDP drilling data to obtain a view of the geologic history of the area. Basement is attenuated lower Paleozoic continental crust with Lower Jurassic diabase dikes. Post-Paleozoic sedimentation began in Triassic(?)-Jurassic(?) rift grabens, probably followed by Early to Middle Jurassic terrestrial to shallow-marine elastics and Late Jurassic shallow and deep-water carbonates. By earliest Cretaceous time, much of the southeastern Gulf was a deep seaway between the carbonate platforms of Florida and Yucatan. There was a steady background of pelagic carbonate sedimentation. In addition, sedimentation was influ- enced by input from the adjacent platforms until middle Cretaceous, by input from the approaching Cuban island arc in Late Cretaceous-Eocene, and by steadily increasing submarine erosion from late Eocene on. According to sediment facies and position relative to the Florida-Bahama platform, the deep southeastern Gulf, where basin Sites 535 and 540 are situated, is an extension of the Camajuani Zone in northern Cuba. The basement Sites 537 and 538 are an extension of the lower Paleozoic (Pan-African) metamorphic terrane in northern Florida. All three basement sites (536, 537, and 538) were probably attached to Yucatan since the Early Cretaceous, as indicated by sediment facies and seismic stratigraphy. Tectonic movements included (Triassic-Jurassic?) faulting during the rift stage, Late Jurassic-Cretaceous block faulting probably related to transcurrent movements between the Gulf and the Atlantic, and finally, faulting along the north slope of Cuba during the Late Cretaceous through Eocene interval.
INTRODUCTION The seismic data were used to develop a preliminary seismic stratigraphic framework and to choose the drill- The purpose of this chapter is to summarize the re- ing sites for Leg 77 (Fig. 2). Data for the lines desig- gional geologic history of the southeastern Gulf of Mex- nated GT-2 and GT-3 were collected in 1977 and 1978 as ico on the basis of the interpretation of multifold seis- part of a regional study sponsored by industry. Data for mic data, collected by the University of Texas Institute the lines designated SF were collected late in 1980 as for Geophysics, and the results of DSDP Leg 77. First, part of an IPOD site survey for Leg 77. Drilling has we discuss the seismic stratigraphy of the area and how added ground data in several strategic locations. Shal- it ties in with the drilling results. The resulting frame- low basement was the objective of drilling at the three work is used to develop a model for the geologic history western holes (536, 537, and 538A), and penetration of of the southeastern Gulf, which is then compared with a thick Mesozoic basin fill was the objective at the east- the known geologic history of surrounding provinces: ern holes (535 and 540) (Fig. 2). All these holes are pro- Florida, Bahamas, Cuba, Yucatan, and the North At- jected into schematic cross-sections in Figure 3. Details lantic. Finally, we discuss the implication of seismic sur- of the drilling results are presented in the site chapters veys and drilling for tectonic models of the early open- and specialty chapters of this volume. ing of the Gulf of Mexico and the collision of the Cu- The seismic data and drilling results allow, for the ban arc with the North American Plate. first time, a preliminary interpretation of the seismic stratigraphy, structural setting, and geologic history of SEISMIC STRATIGRAPHIC FRAMEWORK the southeastern Gulf. This area presents the best op- The University of Texas Institute for Geophysics portunity to study the Mesozoic history because a thick (UTIG) has collected an extensive grid of multifold seis- section of older rocks lies at shallow depth or is exposed mic data in the southeastern Gulf of Mexico during the by erosion at the seafloor. The lack of overburden im- past few years. The study area lies north of Cuba be- proves seismic resolution in the older sedimentary se- tween the Campeche and Florida escarpments (Fig. 1). quences and puts them within reach of the Glomar Here the seafloor is shallower than the deep Gulf to the Challenger. north, and is characterized by erosional channels and The deep southeastern Gulf is underlain by rifted and large knolls. attenuated "transitional" crust covered by a thick sedi- mentary section of pre-middle Cretaceous rocks (Fig. 3). The Late Cretaceous-Cenozoic cover is relatively thin Buffler, R. T., Schlager, W., et al., Init. Repts. DSDP, 77: Washington (U.S. Govt. over most of the area, although it thickens considerably Printing Office). 2 Present address: Texaco, USA, 3350 Wilshire Blvd., Los Angeles, California 90010. to the south, toward Cuba. The pre-middle Cretaceous * Present address: Texaco, Inc., Box 60252, New Orleans, Louisiana 70160. section probably reflects an overall transition upward
715 -J
22° N 87° 86° 85° 84° 83° W Figure 1. Map of southeastern Gulf of Mexico showing major topographic features. Water depths in meters. GEOLOGIC HISTORY
24° N
84° W Figure 2. Map of study area, southeastern Gulf of Mexico, showing DSDP sites and locations of UTIG multifold seismic lines; heaviest lines indi- cate portions of seismic lines illustrated in this chapter. Water depths in meters. from nonmarine to shallow marine and then deep ma- tiary sediments. On the west and east are the Campeche rine deposits as the basin subsided. and Florida escarpments, respectively, which represent A generalized tectonic setting for the pre-middle Cre- the somewhat eroded flanks of the Cretaceous carbon- taceous rocks is illustrated in Figure 4. The area to the ate platforms that rim the deep Gulf. The seismic data north is characterized by a relatively thick section filling establish regional orthogonal fault trends in NW-SE and a subdued basement topography. Just to the south is a NE-SW directions. large area characterized by tilted fault blocks with adja- The sedimentary sequences overlying basement are cent basins filled in with syn-rift sequences. The south- grouped into five main subdivisions: (1) a Late Triassic- eastern part of the area in the vicinity of Jordan Knoll is Early Jurassic(?) rift basin (TJ); (2) a widespread Juras- characterized again by subdued basement relief and a sic(?) nonmarine to shallow marine unit (Jl); (3) a more thick Mesozoic section. It is flanked on the west by a restricted Late Jurassic(?) shallow to deep marine unit large half-graben or graben trending northwest-south- (J2); (4) a widespread Early Cretaceous unit drilled on east. West of the graben is a large stable block with a Leg 77 (EK); and (5) a Late Cretaceous-Cenozoic unit thick older Mesozoic(?) section that is undeformed. Cut- (KC). Table 1 describes each of these major subdivi- ting across the northern part of this area is an older sions. The Early Cretaceous and the Late Cretaceous- buried graben system trending northeast-southwest. The Cenozoic subdivisions have been the subjects of more northern margin of this stable block is faulted in such a detailed studies (Phair, in press; Angstadt, 1983). way as to influence the topography of the seafloor. The The Early Cretaceous (EK) rocks are divided into four entire area in front of Cuba has been depressed consid- main mapping units, as defined in the vicinity of Sites erably to form a foredeep with over 4000 m of lower Ter- 535 and 540 along line SF-15 (Fig. 5). The mapping
717 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR
West 538 East
537 MCU
Rifted continental crust
.West East -20 km
Campeche Escarpment
- talus
Rift bas n?-~ ' Meso; TJ platform?
" "' '''"/V/o 718 GEOLOGIC HISTORY 24° N Catoche Tongue 23' 86 84° W Figure 4. Sketch map showing generalized tectonic setting of study area (water depths in meters). Bold numbers designate the following tectonic provinces: (1) thick sediment section filling subdued basement topography; (2) tilted fault blocks, tectonic depressions filled with synrift sedi- ments; (3) thick Mesozoic sediments overlying basement with subdued relief; (4) half-graben or graben; (5) untilted block with undeformed pre- Cretaceous sediments; (6) northeast-southwest graben; (7) foredeep of Cuban fold belt. In the Middle Jurassic, the area apparently was a pression trending north-south also formed during the broad depression characterized by large basement up- later part of Jl time (Fig. 3). The western margin of this lifts and basins. A widespread sedimentary unit, Jl, is trough may represent relief on an east-facing shelf mar- filling this depression. It probably consists of alluvial gin, since the seismic surveys indicate a broad unde- fans, lacustrine deposits, and volcanics. During the later formed platform in the entire southwestern part of the part of this period, the central position of the southeast- area (Figs. 3 and 4). ern Gulf was subjected to extensive block faulting (num- After deposition and deformation of the Jl unit, there ber 2 in Fig. 4). Older sequences were rotated to form was a broad marine transgression into the area. This half-grabens, and (nonmarine?) rift sediments filled the probably took place in the Late Jurassic along with the intervening basins (Fig. 15). transgression of the Gulf basin to the north. Marine de- Intensive block faulting has been restricted to the cen- posits are inferred for J2 from the seismic similarity to tral part of the southeastern Gulf. Farther north, the Jl the drilled Early Cretaceous sequences above. Low-re- unit onlaps basement and appears to pinch out toward lief shelf margins in the area, interpreted from the seis- the outer high at the inferred ocean crust-transitional mic data, suggest that shallow-water carbonate platforms crust boundary (Fig. 16). This suggests that it is older were present that graded laterally into slightly deeper ma- than the formation of the ocean crust. rine basins. These margins probably were controlled by An early transgression may have reached the area from paleotopography on the underlying, deformed Jl unit. the south late in Jl time. This conclusion is based on the In some places the J2 unit, too, is deformed, indicating shallow-marine dolomite drilled at Site 536, which may that movement of some blocks continued into the Late correspond to the upper part of Unit Jl. A broad de- Jurassic. Before the end of the Jurassic, therefore, a 719 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR Tkble 1. Major seismic subdivisions, southeastern Gulf of Mexico. Seismic facies and Unit Age regional variation Equivalent in borehole Interpretation KC Late Cretaceous-Cenozoic In the south, the lower part (Cretaceous- Drilled thin distal part of unit at Eocene-Recent consists of pelagic on basis of DSDP Holes Eocene) forms thick wedges, coming Sites 540 and 97, approxi- drape in southern and central 97, 536, 537, 538A, and from Cuba, characterized by continu- mately 270 m of carbonate part of area. Onlapped by 540 ous to discontinuous, divergent ooze, chalk, and marl. distal turbidites of Gulf basin facies. Upper part (Eocene-Recent) is Lower part disturbed by to the north. Cenomanian- a thin blanket, with many uncon- creep and slumping. Lower Eocene forms a thick wedge formities. Unit thins to north in 56 m are gravity-flow de- of turbidites and gravity-flow drilling area and then thickens again posits. deposits in foredeep of Cuban into Gulf basin farther to the north. orogen. Local carbonate Four mapping units (Tkble 3). debris shed from Florida Scarp (Angstadt, 1983). EK Early Cretaceous on basis Widespread throughout northern and Drilled almost entire unit in Deep-water carbonate deposition of DSDP Holes 535, eastern part of area. Up to 2 km Holes 535 and 540; approxi- in northern and eastern part 536, 537, 538A, and 540 thick. Thickens to east along base of mately 750 m of pelagic of study area. Main source Florida Escarpment. Thin or absent limestone-marly limestone was Florida platform to east to south and west owing to both cycles with episodic interbed- and planktonic carbonate nondeposition on high-standing areas dings of shallow-water lime production (Phair, in press). and post-mid-Cretaceous erosion. sand and mud. Thin equiva- Subdivided into several sequences lents drilled at Holes 537 and four mapping units (Tkble 2). and 538A (elastics and Relatively uniform parallel continu- shallow-water limestones ous facies over most of the area. overlain by pelagic lime- More discontinuous and hummocky stones) and at Hole 536 along Florida Escarpment. Promi- (talus deposits). nent N-S hummocky facies in middle unit. J2 Probably Late Jurassic on Uniform, variable-amplitude, continuous Not reached Seismic character similar to basis of position below reflections. Widespread over most of Cretaceous above, suggesting Earliest Cretaceous area. Deformed in depressions be- deep-marine deposition in drilled in Hole 535 tween horsts. Absent on high-stand- central part of area. Possible ing blocks to west. Slightly deformed low-relief shelf margins in in places owing to later tectonic places suggest transition to movement. shallow-marine around pe- riphery of area. Probably represents major marine transgression and establish- ment of seaway in area (Phair and Buffler, 1983). Jl Inferred to be Jurassic, Widespread unit in south with relatively May have just reached top of Seismic character and setting to possibly Middle Juras- uniform thickness (several km) and unit in bottom of Hole 536. north suggests non-marine sic, on basis of superpo- seismic character (high-amplitude, Shallow-water dolomite, synrift sediments (alluvial sition discontinuous). Onlaps broad base- Jurassic or Permian on basis fans, lacustrine deposits, ment highs. Undeformed in south of 87/86 Sr ratio. volcanics, evaporites). Upper except down-dropped along promi- part of unit in south may be nent NW-SE graben system and shallow-marine platform. along broad trough north of Cuba. Lower part to south may be To north unit fills half-grabens be- nonmarine (Phair and Buf- tween tilted fault blocks. fler, 1983). TJ Inferred to be Late Trias- Southeast-dipping parallel reflections Not reached Inferred to represent equivalent sic-Early Jurassic on filling NE-SW-trending graben of Late Triassic-Early Jurassic basis of setting system. Onlaps basement. Truncated rift sequences observed else- by prominent unconformity. where around margins of North Atlantic and Gulf of Mexico. Probably nonmarine sediments and volcanics (Phair and Buffler, 1983). Basement early Paleozoic (500 Ma) Acoustic basement over much of the Drilled phyllite in Hole 537 and Top of rifted and attenuated metamorphic rocks area. In places contains some low- gneiss-amphibolite in Hole (transitional) crust (Buffler et intruded by early Meso- amplitude reflections. Occurs as 538A intruded by basic dikes al., 1981; Ibrahim et al., zoic (160-190 Ma) dikes broad uplifts and basins to south and and sills. 1981). Probably mostly early and sills north. High-relief tilted fault-blocks Paleozoic (Pan-Africa) meta- in central part of area. morphic rocks, possibly with Paleozoic sedimentary rock in places. Dissected by Jurassic intrusives and extrusives. deep marine seaway had been established in the south- some extent, 536 can be firmly tied to the seismic stra- eastern Gulf, a setting that persisted throughout the tigraphy of the area. Sites 537 and 538, with condensed Early Cretaceous, as discussed next. sections on high-standing fault blocks, contributed sig- nificantly to the interpretation of tectonic movements Early Cretaceous (EK Sequences) and the nature of basement. This is the first time-interval for which boreholes di- Throughout the Early Cretaceous, the southeastern rectly aid seismic interpretation. Sites 535, 540, and, to Gulf was strongly influenced by the rim of carbonate 720 GEOLOGIC HISTORY Southwest Northeast 10 .km. 535 540 Figure 5. Depth section of seismic Line SF-15, showing Early Cretaceous seismic units EK1-4 near Holes 535 and 540. KC = Late Cretaceous-Ceno- zoic seismic units. platforms. The effect on basin sedimentation increased Sites 537 and 538 present one of the most controver- as the platforms rose higher and their slopes steepened. sial sedimentologic problems of Leg 77: the depositional Seismic sequences of the Berriasian-Hauterivian inter- environment of the Early Cretaceous limestones. We will val vary little in thickness throughout the area, and cor- briefly review this topic because it bears on the timing of respond to purely pelagic limestones with slow rates of block faulting and the relation of the basement sites to deposition at Site 535. From late Hauterivian on, the in- the Yucatan platform. In the site chapters, it was left un- creasing sediment input from the rapidly rising Florida decided whether the Berriasian-Valangiiiian limestones platform is evident (compare Fig. 6 with Figs. 7 and 8). were platform talus, in situ limestone caps (atolls), or Seismic units thicken toward the Florida Escarpment, faulted parts of a much larger Cretaceous carbonate and more shallow-water debris appears in the cores. Sim- platform. ilar but narrower tongues of talus rim the Campeche Es- The fundamental question is whether the limestones carpment, and were drilled at Site 536. are neritic platform deposits or platform talus shed into The high sediment input from the platforms, clearly deep water (Fig. 17). Compared with those at Site 536, seen in the geometry of seismic sequences, reached the which we consider proven talus, the limestones at Sites basin Sites 535 and 540 only episodically. Material de- 537 and 538 lack almost all features that supported posited at these sites was predominantly pelagic carbon- the talus interpretation at 536. We observed no interbeds ate, showing a very persistent alternation of dark, lami- of pelagic chalk, no pelagic internal sediment in neritic nated layers rich in organic matter with light, bioturbat- material, no talus texture. We did find plankton—that ed layers rich in carbonate. Similar cycles characterize is, calpionellids and globigerinids—mixed with the ne- the coeval deposition in the Atlantic (Arthur and Nat- ritic biota. However, there are several well-documented land, 1979). Cotillon and Rio (this volume) relate the examples of the. association of calpionellids and neritic cycles at Sites 535 and 540 to climatic variations driven benthos in open-marine shallow-water limestones (Blox- by the Earth's orbital perturbations. som, 1972; Baudrimont and Cussey, 1977; Finneran et Sites 537 and 538 on fault blocks in the western part al., 1982). At Site 536, the setting at the foot of Cam- of the study area serve only indirectly as calibration points peche Bank is a strong argument for the interpretation for the seismic grid. They record a Berriasian transgres- as talus (see site chapter, Site 536, this volume). At Sites sion on basement, confirming the conclusions, from seis- 537 and 538, the setting argues against the talus hypoth- mics, that the western portions of the southeastern Gulf esis because there is no evidence for Neocomian plat- stood high during the Jurassic-Early Cretaceous and that forms nearby. Furthermore, the platforms that did exist thick deep-water sections were confined to a 100-km- had a rather modest relief and shed little talus, quite in wide seaway in the center (Figs. 6 and 7). contrast to their Aptian-Albian counterparts. The simi- 721 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR Seismic Unit EK1 Figure 6. Isochron map of Early Cretaceous seismic unit EKl (two-way traveltime in seconds). Water depth in meters (heavy lines). larity to Site 536, in terms of drilling rate and low recov- the site was still part of a larger drainage pattern, not an ery, is also no argument for talus. DSDP teams on the isolated high as it is now. Because the limestones overlie Challenger had similar recoveries (2-14%) in genuine the elastics with no discernible age difference (see site platform carbonates (e.g., Campeche Bank: Worzel, Bry- chapter, Site 537, and Oertli, both this volume), we be- ant, et al., 1973; Ita Matai seamount in the Pacific: lieve that they, too, were part of a more extensive car- Heezen, McGregor, et al., 1973). The only remaining ar- bonate platform, not an isolated atoll. At Site 537 this gument in favor of talus is the interpretation of certain platform developed over a blanket of siliclastics, where- benthic foraminifers in the limestones as bathyal (Sliter as at Site 538 it has prograded over deep-water chalk. and Premoli Silva, this volume). We consider this argu- The implications of the platform hypothesis for the ment inconclusive because of the uncertainties in the pa- question of Cretaceous faulting will be discussed later. leo-ecologic interpretation of pre-Tertiary assemblages. Taken overall, the evidence for in situ limestones far out- Late Cretaceous Hiatus and Mid-Cretaceous weighs the arguments against it. Unconformity (MCU) The next question is whether these limestones formed At all DSDP sites in the southeastern Gulf of Mexi- two separate, atoll-like caps on high fault blocks or are co, Late Cretaceous sections are very thin or missing. In segments of a larger platform faulted after deposition. the basement Holes 536, 537, and 538A, the Late Creta- The sequence at Site 537 supports the latter interpreta- ceous is represented by a few meters of carbonate ooze tion. The section consists of an Early Cretaceous suc- and volcanic ash with numerous hiatuses, but with gen- cession of alluvial elastics, littoral elastics and dolo- erally well preserved calcareous biota. In the basin Holes mites, and neritic limestones. There is a mismatch in 540 and 97 (Worzel, Bryant, et al., 1973), the Cenoma- composition between the basement (i.e., phyllite) and nian through early Paleocene sequence is represented by the overlying elastics, which are derived from acidic ig- about 50 m of sediment gravity flows. neous rocks (see site chapter, Site 537, and McBride, We propose a combination of low sediment input, cur- both in this volume). Thus, at the time of transgression, rent scouring, and slumping on tectonically steepened 722 GEOLOGIC HISTORY Seismic Unit EK2 100 km 85° 84° 83°W Figure 7. Isochron map of Early Cretaceous seismic unit EK2 (two-way traveltime in seconds). Water depth in meters (heavy dashed lines). slopes to explain the extreme reduction of the Late Cre- val we find increasing evidence for creep and slumping: taceous section. Sediment input was low because the inclined bedding, folds, intraclasts, and occasional mud- carbonate platforms rimming the Gulf had been drowned flows with exotic limestone clasts. Unusually high sedi- during the Albian-Cenomanian (Arthur and Schlanger, ment accumulation rates of 120 m/Ma fit well with the 1979; Schlager, 1981); this greatly reduced the input of lithologic evidence for reworking and slumping. The Pa- shallow-water carbonate. Furthermore, carbonate dis- leocene-late Oligocene sequence resembles the late Albi- solution at relatively shallow depth may have reduced an insofar as it also contains slump folds, inclined and planktonic carbonate sedimentation in the Aptian through wavy layering, and flow structures. The 56 m of sediment Campanian interval (Thierstein, 1979). Episodic erosion gravity flows between the early Cenomanian and Paleo- is indicated by hiatuses, hardgrounds, and reworked bi- cene horizons contains a normal stratigraphic succession ota in the holes on basement highs. In these locations, of early or middle Cenomanian, Maestrichtian, and late contour currents are a likely mechanism of erosion. A Paleocene biota. This section can be interpreted either juxtaposition of the Cuban arc approaching from the as a complex of mudflows and turbidites emplaced dur- south and the southern margin of the Bahama-Florida ing the Paleogene collision of Cuba and Florida or as a platform as postulated by Malfait and Dinkelman (1972) series of Cenomanian through Paleocene sediment grav- and Sykes et al. (1982) may have channeled and intensi- ity flows. In the latter case, tectonic movements or ero- fied deep-sea circulation in the area. sion during falls of sea level are possible causes. Further The effects of slow sedimentation and current scour- sliding and creep occurred in Eocene-late Oligocene. ing notwithstanding, we feel that tectonics played an im- An overview of the regional picture, including Cuba portant role in reducing the Late Cretaceous section. We and the Cat Gap area in the Atlantic, supports the view think that slopes were tilted and sediment slid off or was of intermittent tectonic activity throughout the Albian- eroded through the increased vigor of sediment gravity Eocene interval. In Cuba, the northern miogeosynclinal flows. The record of Site 540 is particularly interesting belt shows numerous gaps in the Late Cretaceous sec- in this respect. Throughout the middle Albian, the sec- tions (Pardo, 1975; Pszczolkowski, 1978). At least some tion is practically undisturbed. In the late Albian inter- of these are angular unconformities (Khudoley and Mey- 723 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR Seismic Unit EK3 Figure 8. Isochron map of Early Cretaceous seismic unit EK3 (two-way traveltime in seconds). Water depth in meters (heavy dashed lines). Table 2. Early Cretaceous seismic units, southeastern Gulf of Mexico. Seismic facies and Unit Age regional variation Equivalent in borehole Interpretation EK4 early Cenomanian- Parallel, continuous reflectors of Site 540: 417 m of hemipelagic late Albian moderately high amplitude; limestone-marly limestone cycles toward Florida Escarpment, with some shallow-water detritus reflectors become discontinu- ous and unit thickens rap- idly; restricted to eastern EK2-4: Mainly pelagic carbonates in part of study area. SW, grading rapidly into hemipe- EK3 early Albian-late Parallel-subparallel, continuous, Site 535: 233 m of subdued cycles of lagic deposits with much rede- Albian of moderate amplitude, limestone and marly limestone, posited platform material toward grading westward into hum- frequent interbeds of shallow- Florida Escarpment; rates and mocky clinoforms, suggest- water calcarenites (and mud?). loci of platform input changed ing westward shift of depo- Equivalent at Site 536 is plat- with time, creating sequence center with time; absent in form talus; at 537 and 538, boundaries. extreme W and largely pelagic chalk and limestone. eroded near Cuba. EK2 late Aptian-late Parallel, continuous, and high- Site 535: 80 m of cyclic alternation Hauterivian amplitude in SW; changes to of pelagic white limestone and (several hia- low-amplitude, discontinu- marly limestone rich in organic tuses) ous toward Florida Escarp- matter. Equivalent at Sites 537 ment; rapid thickening in and 538 is condensed section of same direction; thin or pelagic chalk and limestone. absent in SW. EK1 Hauterivian-late Parallel, continuous, with mod- Site 535: 248 m of pelagic limestone Mainly pelagic carbonates; minor Berriasian erate amplitude; thickens at in cyclic alternation with marly sediment contribution from low- foot of northern Florida limestone rich in organic matter; relief platform in NE (Florida); Escarpment; thin or absent hardgrounds (increasing in fre- in W affected by synsedimentary in SW; thickness highly quency downhole). block faulting. variable in block-faulted area. Note: Most of this material is after Phair, in press. 724 GEOLOGIC HISTORY 5 (Ewing, Worzel, et al., 1969) and Holes 99, 100, and Seismic Unit 101 (Hollister, Ewing, et al., 1972) show stratigraphic KC1 gaps in the Albian-Paleogene interval, and reduced Late Cretaceous sections with pebbly mudstones similar to those at Site 540 were recovered at Sites 4 and 5. The "mid-Cretaceous unconformity" or MCU is a seismo-stratigraphic marker of basin-wide significance in the Gulf of Mexico. Around the margins of the Gulf, it is an unconformity characterized by truncation below / ) and onlap above. In the center of the basin, it becomes " r'Pinardel • a conformity marked by a high-amplitude reflector. It has '-fl/o Knoll been traced from DSDP Site 97 throughout the south- eastern Gulf of Mexico to the deep western basin (Buf- fler et al., 1980). In the following we will discuss both timing and origin of this seismic reflector. Figure 9. Isochron map of Late Cretaceous-Cenozoic seismic unit KC1 The supposition that the reflector is mid-Cretaceous (two-way traveltime in seconds). Water depth in meters (heavy has been tentative from the outset. It was based on the dashed lines). reflector's position very near Cenomanian deep-water sed- iments at the bottom of DSDP Hole 97, and on a seis- mic tie to wells in the northeastern Gulf and on Cam- Seismic Unit peche Bank (Buffler et al., 1980; Addy and Buffler, in KC2 press). The correlation at Site 97 is weak because the hole has only been spot-cored. Hole 540, only 9 km away, penetrated an almost identical seismic section with continuous core recovery, but the correlation remains ambiguous because of the Late Cretaceous hiatus al- ready discussed. The MCU is also tied to the Campeche and Florida banks, where a strong case can be made for the reflector being mid-Cretaceous (Buffler and others, 1980). In interpreting the MCU reflector for this chapter, Buffler, Angstadt, and Phair emphasize the role of ero- sion by deep-sea currents during the Cenomanian low- stand of sea level postulated by Vail and others, 1977. Figure 10. Isochron map of Late Cretaceous-Cenozoic seismic unit Schlager sees platform drowning during rises of sea level KC2 (two-way traveltime in seconds). Water depth in meters (heavy as the prime cause of the MCU. In his opinion, the dashed lines). MCU is a stratigraphic turning point caused by decima- tion of the carbonate platforms around the Gulf and the concomitant change in sediment input and dispersal. 2 Rapid rises of sea level, possibly coupled with adverse Hole V° Seismic Unit : KC3 environmental factors, caused a worldwide crisis of plat- forms in the mid-Cretaceous (Schlager, 1981). During this time, some of the platforms around the Gulf were completely drowned (e.g., Jordan Knoll, Edwards Shelf, Golden Lane) while others were forced to step back to a more landward position (e.g., Campeche and Florida platforms). This drastically reduced sediment input into the Gulf and shifted the loci of deposition to siliciclastic entry points: first, to the Mexican margins in the west (Shaub et al., in press), and later, after the development of the Mississippi drainage, to the northwest and north. We all agree that both erosion and changes in sedimen- tation occurred in the middle Cretaceous, and that the Figure 11. Isochron map of Late Cretaceous-Cenozoic seismic unit MCU changes its character throughout the basin. In KC3 (two-way traveltime in seconds). Water depth in meters (heavy some places it represents a conformable boundary be- dashed lines). tween carbonate-rich sediments below and carbonate- poor sediments above, in others the onlap of siliclastics erhoff, 1971). More or less continuous Albian-Eocene on inactive carbonate aprons around the drowned plat- compressional tectonism has been postulated by Shein forms, and in still others an erosional unconformity where and others (1978) for Cuba. In the Cat Gap area, east of the Cretaceous deep-sea carbonates have been dissolved the Bahamas, Ewing, Worzel, and others (1969) report and scoured before the siliciclastics arrived. In the south- post-Neocomian tectonic movement. DSDP Holes 4 and eastern Gulf, tectonic décollement and slumping, along 725 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR Northwest Southeast 25 km Vertical exaggeration = 16x Figure 12. Regional northwest-southeast cross-section (heavy bar on inset map) based on depth-converted seismic profiles. Note southward thicken- ing of seismic units KCl and KC2 in the Cuban foredeep and localized erosion in pelagic cover (KC3 and KC4). Table 3. Late Cretaceous-Cenozoic seismic units in southeastern Gulf of Mexico. Seismic fades and Unit Age regional variation Equivalent in Borehole Interpretation KC4 Recent-middle Miocene Parallel, continuous reflection Site 540: S2 m of calcareous ooze Pelagic drape, partly eroded by with numerous small and chalk turbidity currents and contour unconformities; forms currents discontinuous drape. KC3 early middle Miocene- Parallel, continuous grading Site 540: 191 m of chalk and Mostly pelagic drape, complicated late Eocene into discontinuous but not marly chalk deformed by by slumping on (tectonically?) divergent fades near Cuba; sliding (inclined bedding, over-steepened slopes reflection continuity also folds, microfaults, deformed degrading to SW owing to burrows) slumping and valley- cutting. KC2 early late Eocene-late Diverging, continuous, thick- Site 540: 29 m of alternating Mainly turbidites in the south Paleocene ening toward Cuba, hum- chalk and dark marly lime- (Cuba), increasing pelagic mocky along base of stone, deformed by creep or component to north Florida Escarpment; top slumping. Top boundary boundary is major uncon- corresponds to late Eocene formity with considerable hiatus. relief and consistent trun- cation. KCl late Paleocene-mid- Strongly diverging toward Site 540: 56 m of sediment Mainly slumps and sediment Cenomanian Cuba, with concomitant gravity flows (pebbly chalks gravity-flow shed from Cuba, decrease in continuity; overlain by graded conglom- with minor source along minor thickening and erates and sandstones), possi- Florida Escarpment decrease in continuity bly thin interbeds of hemipe- toward Florida Escarp- lagic limestone. Sedimenta- ment. tion discontinuous; major hiatuses. Section represents only small part of correlative seismic unit. Note: Most of this material is after Angstadt, 1983. with block faulting, have modified the reflector even geometry of the coeval seismic units (Figs. 9, 10, and further and merged it with younger reflectors, as at Site 12). Thickness of KCl increases five-fold between Jord- 540. an Knoll and the north slope of Cuba. Isochrons closely parallel the slope, suggesting a line source rather than a Late Cretaceous-Cenozoic (KC Sequence) single-canyon point-source of sediment gravity flows. After the MCU turning point, sediment input into Thickening of KC2 is more gradual, and input is con- the southeastern Gulf was no longer controlled by the centrated in the southwest, suggesting a major canyon- Florida and Yucatan carbonate platforms. Instead, sedi- fan system in this area. Seismic facies indicate a shift ment was shed mainly from the advancing Cuban island from mainly slumps and debris flows in the thickest por- arc during the Late Cretaceous-Eocene interval (KCl tion of KCl to more widely spread turbidites in KC2. In and KC2, Fig. 12). At mid-Eocene time, the northward the north, protected by Jordan Knoll, a rapid transition thrust of Cuba ceased (Sykes et al., 1982; Pindell and occurs into the predominantly pelagic carbonates cored Dewey, 1982). The southern sediment source lost its dom- at Site 540. Other effects of the Cuban orogeny are re- inant position, and pelagic sedimentation prevailed. verse faults and monoclinal folds along the north slope The filling of the Cuban foredeep during the Late of Cuba and synsedimentary down-to-south faults in Cretaceous-early Eocene is reflected dramatically in the KCl along the northern flank of the foredeep (Fig. 12). 726 GEOLOGIC HISTORY The Eocene unconformity on top of KC2 represents cutting seismic units KC1 and 2 in some profiles (Ang- another major change in depositional regime. By that stadt, 1983). time, Cuba was firmly welded to Florida, and flysch In summary, we conclude that faulting persisted, at sedimentation in the foredeep had ended (Angstadt et least episodically, until the Paleogene. It may have start- al., 1983). Seismic units KC3 and 4 in the southeastern ed during a Jurassic rift stage, been rejuvenated during Gulf are composed of rather evenly spread pelagic de- the Tithonian-Neocomian reorganization of the Gulf, posits modified by erosion and redeposition from con- and continued in the Cuban (Laramide) orogeny. tour currents or turbidity currents. Coring in Hole 540 recovered carbonate ooze, marl, and chalk, with hiatus- Relations to Florida, Cuba, and Yucatan es becoming more abundant upward in the section (Wat- The DSDP sites in the southeastern Gulf are pres- kins, this volume). A major hiatus between late Oligo- ently surrounded by three land masses whose relations cene and middle Miocene deposits in Hole 540 corre- to one another and to the Gulf of Mexico are a matter sponds to the boundary of Units KC3 and 4. From late of considerable debate. The results of Leg 77, along Eocene on, turbidity currents seem to have followed a with the extensive seismic surveys and new literature, similar pathway along a northwest- or west-dipping val- lead to certain correlations which we will present and ley with tributaries on the flanks of Cuba and Florida. discuss in the following section. Erosion in these valleys caused several large slides along their flanks. Florida vs. Basin Sites We contend that Sites 97, 535, and 540 were attached Timing of Block Faulting to the Florida carbonate platform from Neocomian time Passive-margin evolution calls for intensive block fault- on. This statement is based on the abundance of shal- ing in the rift stage, followed by a drift stage with steady low-water carbonate detritus in the Cretaceous sections subsidence and only minor rotational movements of the of Sites 535 and 540 and on the geometry of the seismic blocks. The southeastern Gulf has not followed this pat- sequences containing this detritus. Invariably, these se- tern entirely. The boreholes indicate that faulting con- quences thicken toward the Florida Escarpment, with tinued (or was reactivated) throughout Cretaceous and internal reflections becoming more irregular and less con- possibly Eocene time, well past even the drift stage of tinuous, suggesting increasing volumes of slumps and the margins. debris flows (Figs. 7, 8, and 14). A comparison of Sites 537 and 538, located on two horsts 30 km apart, yields evidence for Early Cretaceous Camajuani Zone in Cuba vs. Basin Sites faulting (Fig. 18). We assume Berriasian faulting to ac- The stratigraphic sequence at the basin Sites 535 and count for the different lithologies of this interval in Holes 540 we interpret as the northwestern extension of the 537 and 538A. Post-Valanginian faulting at Sites 537 Camajuani Zone (Pszczolkowski, 1982) in central Cu- and 538 is required to explain their present positions ba, which was deformed and uplifted when the Cuban 1500 to 2000 m above the Cretaceous on the adjacent island arc collided with the Florida-Bahama region. Pa- downthrown blocks. (If one assumes that the Valangini- leogeography, sediment facies, and structure support this an limestone caps were not part of a larger platform but correlation (Figs. 19 and 20). Paleogeographically, the formed as two separate atolls, post-Valanginian faulting basin sites are part of a (Jurassic?)-Cretaceous deep-wa- must nonetheless be postulated because there was no rec- ter area. To the northeast lies a carbonate platform that ognizable difference in water depths at the two sites in was drowned in the Albian or Cenomanian; still farther the Valanginian, whereas the present elevations of their northeast follows the Florida platform, where carbon- Valanginian limestones differ by 228 m.) The post-Val- ate deposition persists to this day. A very similar succes- anginian throw along the faults seems to have increased sion of facies belts is observed in central Cuba (Pardo, gradually. Sand-size neritic biota scattered throughout 1975; Pszczolkowski, 1982): (1) Camajuani Zone, with the deep-water assemblage (see Sliter and Premoli Silva, a continuous deep-water sequence; (2) Cayo Coco Zone this volume) suggests that some turbidity currents swept (and probably Old Bahama Channel: M. Ball, pers. over these highs as late as Campanian or Maestrichtian. comm., 1983), with a carbonate platform up to Aptian, Clay mineralogy indicates intermittent tectonic activity overlain by an Albian-Eocene deep-water sequence; and around the margins of the southeastern Gulf during the (3) continuous Mesozoic-Cenozoic shallow-water car- Barremian-Albian (Debrabant et al., this volume). bonate in the Yaguajay Zone (Fig. 20). Eocene movement is suggested by a slump in Hole The Cretaceous of the basin sites is lithologically sim- 537. A 9-m interval of Paleocene chalk with slump folds ilar to that of the Camajuani (equals Las Villas Zone). and tilted bedding capped by a hardground lies sand- Pardo (1975) describes the Tithonian-Türonian in this wiched between flat-lying Eocene chalks. We believe zone as a well-bedded sequence of limestones with yel- that tectonic tilting rather than sediment overloading of low-orange weathering, wavy laminae, and marly shales the slope caused the slump, because the sedimentation (probably equivalents of the marls, rich in organic mat- at the site was slow and frequently interrupted (hard- ter, at Sites 535 and 540). Black chert and turbidites with grounds), which makes failure as a result of overloading neritic material occur in the Aptian-Turonian interval; unlikely. there are also hiatuses in the Coniacian through Paleo- Faulting along the north slope of Cuba seems to have cene, and poorly preserved ammonites are common, much continued into the late Eocene, as suggested by faults as at Sites 535 and 540. 727 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR North Tilted basment blocks Figure 13. Portions of seismic line GT3-75 near the base of the Campeche Escarpment, showing the regional setting of Sites 536 and 537. Site 536 is on a large high-standing basement block. Just to the south is a rift basin truncated and onlapped by the Jl unit. To the north is a lower area of fault blocks. Older sedimentary sequences have been tilted and younger basins filled with synrift sediments. Site 537 is on the top of one of the tilted blocks. Drilling there recovered 500( ± )-Ma-old phyllite. On the basis of stratigraphy and position in the nappes, Whether the similarities are coincidental or reflect Pszczolkowski (1982) correlates the Quinones and the proximity in space remains open. It should be noted, La Esperanza-Rosario zones in western Cuba with the however, that Sites 537 and 538 lie approximately in the Camajuani Zone and the Placetas Zone, respectively, in projected extension of the Placetas Zone (Fig. 19). central Cuba. This correlation implies that neritic-litto- Basement ages of 61 Ma and 180 Ma have been re- ral elastics of the Jurassic Cayetano Group probably un- ported for the Placetas sequence. This is much younger derlay the deep-water carbonates of the Camajuani Zone than the 500 Ma ages found at Sites 537 and 538. How- in central Cuba before the sequence was tectonically ever, the first date was measured on secondary biotite, truncated. and the second date is of questionable validity (Khudo- ley and Meyerhoff, 1971). Placetas Zone in Cuba vs. Sites 537 and 538 From the above comparison of the basin sites with An interesting parallel exists between Sites 537 and the Camajuani-Quinones Zone of Cuba, we postulate a 538 and central Cuba. The Placetas sequence (Pszczol- Jurassic-Neocomian deep-sea connection between the kowski, 1982; equal to Cifuenetes Belt, Pardo, 1975) re- Gulf of Mexico and the Atlantic south of the Bahamas. sembles Sites 537 and 538 in all major features except If one accepts the north-south succession of zones pro- the shallow-water limestones, which are absent in Cuba: posed by Khudoley and Meyerhoff (1971), Pardo (1975), and Pszczolkowski (1982), this seaway consisted of a Placetas Sequence northern margin of carbonate platforms, passing south- Cifuentes Belt (Pszczolkowski, ward into deep-water carbonates (Camajuani Zone) and, (Pardo, 1975) 1982) Site 538 still farther southward, into pelagic shale and chert (Pla- cetas Zone) (Figs. 19 and 20). We believe this seaway Pelagic chert, clay Radiolarite, claystone Pelagic limestone chert (Alb.-Tür.) (Apt.-Cenom.) (Alb.-Cenom.?) connected with the Atlantic either across the southeast- Pelagic limestone Pelagic limestone Neritic limestone ern Bahamas (Schlager et al., 1984) or even south of (Neocomian) (Berrias.-Haut.) (Berrias.-Val.) them. Florida and the northwestern Bahama Banks were Conglomerate Arkosic sandstone Arkosic sandstone welded into a single carbonate platform until mid-Cre- (Neocomian) (Berriasian) (Berriasian) Pink granite, Gneiss, amphibolite taceous time, as indicated by widespread Early Creta- amphibolite with diabase dikes ceous evaporites and by seismic evidence for shallow- water carbonates in the Straits of Florida and the Provi- 728 GEOLOGIC HISTORY early Mesozoic? platform 729 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR Southwest J2 Acoustic basement (rifted continental crust—transitional crust) Figure 14. Portions of regional seismic Line SF-4 showing MCU, Early Cretaceous, and Jurassic units overlying basement. Prominent wedge of sedi- ments along base may correlate with MCU. an limestones at Sites 537 and 538 probably form parts sic formations in western Cuba (Khudoley in Khudoley of a much larger carbonate platform. Evidence for this and Meyerhoff, 1971). platform is seen on seismic lines in an area northwest of the sites and northeast of the Campeche Escarpment. Comparison with North Atlantic We consider it the underpinning of the higher but less Most tectonic reconstructions assume a close connec- extensive Aptian-Albian platform (Fig. 17). tion between the Mesozoic North Atlantic and the Gulf of Mexico (e.g., Buffler et al., 1980, 1981; Klitgord et Stratigraphy Below Hole 535 al., in press). Thus, it is instructive to compare the sedi- The connection between the Camajuani Zone and the mentary sequences in the southeastern Gulf, notably at basin sites allows one to use the sections exposed on Cu- Sites 535 and 540, with the well-documented central North ba to complement the borehole stratigraphy. We pro- Atlantic. Formally named formations have been estab- pose as a working hypothesis that the well-layered seis- lished in the North Atlantic, and have been correlated mic stratigraphic units between 4300 and 5000 m at Site over vast areas (Jansa et al., 1979; Tucholke and Moun- 535 (Fig. 5) are equivalent to the Late Jurassic deep-wa- tain, 1979). The similarity between Gulf and Atlantic ter limestones in western Cuba. In the Rosario Zones, sites is greatest in the Neocomian interval. The two ba- this sequence—the Artemisa Formation—has recently sins followed different lithologic trends from then on. been described in detail by Pszczolkowski (1978). It con- From the Berriasian through Barremian, the lime- sists of 700 to 800 m of Oxfordian-Hauterivian micritic stone-shale cycles of the Blake-Bahama Formation are limestones with pelagic fauna (ammonites, tintinnids, closely matched by the limestone-marl cycles of the same and planktonic foraminifers) and (turbiditic?) interbeds interval at Site 535, both reflecting rhythmic alternation of neritic carbonate material. The deep-water Artemisa of aerobic and anaerobic bottom conditions, probably Formation grades downward into the mainly neritic Ja- coupled with variations in carbonate accumulation. Ear- gua and Francisco formations (shale, with sandstone and ly Cretaceous sedimentation in the southeastern Gulf limestone; Oxfordian), underlain by the shallow-water was more calcareous than in the North Atlantic, and the elastics of the Cayetano Formation (Middle and Early terrigenous component was reduced to a background of Jurassic). The more discontinuous and more steeply clay. We attribute this largely to a difference in hinter- dipping reflectors below 5200 m in Figure 5 may be land. In the Gulf, the rim of carbonate platforms was equivalents of the Cayetano elastics, which are consid- nearly continuous (e.g., Bryant et al., 1969), whereas in ered by some to underlie unconformably the Late Juras- the Atlantic this belt seems to have been breached by av- 730 GEOLOGIC HISTORY Northeast — 1 10 EK2 Acoustic basement (rifted J2 EK1 continental crust—transitional crust) Figure 14. (Continued). enues of major clastic influx, for instance, off Morocco to 2500 m. Thus, either the sites in the Gulf were situ- (Lancelot, Winterer, et al., 1980) and in the northern ated at a depth of 2000 m or less during the Late Creta- part of the North American margin (Jansa et al., 1979). ceous or the Cretaceous Gulf of Mexico was distinctly A drastic difference of facies is apparent in the Barre- more saturated with carbonate than the adjacent North mian-Cenomanian interval, which juxtaposes carbona- Atlantic. Benthic foraminifers provide evidence to sup- ceous shale of the Atlantic Hatteras Formation and thick port the first interpretation (Sliter and Premoli Silva, limestones at Sites 535 and 540. Like the black Hatteras this volume). Shale, the mostly laminated carbonaceous limestones at The early and middle Eocene in the North Atlantic Sites 535 and 540 indicate predominantly anaerobic con- was characterized by abundance of siliceous plankton ditions. However, most of the Atlantic sites lay below (mainly radiolarians) and chert (Bermuda Rise Forma- the CCD, whereas the Gulf sites were at least intermit- tion). In the southeastern Gulf, we notice a "silica spike" tently above it. The facies succession in central Cuba in the same interval, but it is less prominent. At Site suggests that the deep-water limestones of the Cama- 540, occasional chert nodules appear in marly chalks juani Zone pass southward into a deeper basin with dark that otherwise differ little from overlying and underly- shale and chert (Placetas Zone, Pszczolkowski, 1982). ing beds. At Site 538, the middle Eocene sediments con- The Placetas rocks may be a better facies equivalent of tain layers of radiolarian mudstone in chalk. Nowhere the Hatteras Shale than the limestones of Sites 535 and are the siliceous lithologies prominent enough to estab- 540 (Fig. 20). lish a separate unit for them. During the Cenomanian-Paleocene, low productivity Most of the Tertiary in the western North Atlantic is and shallow position of the CCD led to widespread de- represented by hemipelagic muds, strongly reworked and position of pelagic clays in the western North Atlantic redistributed by contour currents (Blake Ridge Forma- (Plantagenet Formation, Jansa et al., 1979). In contrast, tion, Jansa et al., 1979). In the southeastern Gulf, on the southeastern Gulf of Mexico was clearly above the the other hand, marly carbonate ooze prevailed until the level of intensive carbonate dissolution, as indicated by fringes of the Mississippi fan reached the area in the carbonate ooze with very well preserved calcareous nan- Pleistocene. nofossil floras at Sites 537 and 538 and by equally well preserved Campanian nannofossil floras that were found Implications for Tectonic Models of the reworked in Pleistocene mud at Site 535. Thierstein (1979) Gulf of Mexico shows for the Turanian-early Campanian in the North The geologic history we have outlined here needs to Atlantic a sharp decrease in carbonate content, indicat- be compared with existing tectonic models for the early ing the onset of intensive dissolution as shallow as 2000 evolution of the Gulf of Mexico. Most current models 731 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR West .MCU Basement? Figure 15. Line SF-2 crosses the area of block-faulted basement. The graben east of Catoche Knoll is floored with a faulted and deformed Middle Jurassic 732 GEOLOGIC HISTORY East 10 km MCU EK (synrift?) Figure IS. (Continued). to the east or southeast. Thus, the motions of Yucatan al positions of these elements relative to one another re- must have occurred in pre-Cretaceous time. mained unchanged throughout this period. Cuba's posi- In summary, we favor a model that interprets the south- tion relative to the southeastern Gulf became fixed in eastern Gulf as a broad transform boundary during the the late Eocene. rifting and spreading of the western Gulf basin. There 3. Sedimentation in the southeastern Gulf was domi- was no complete separation and formation of ocean crust nated by input from carbonate platforms during the Ear- in the area; it is underlain entirely by attenuated conti- ly Cretaceous, by input from the advancing Cuban arc nental crust or "transitional" crust. The presently mapped from late Cretaceous through late Eocene, and by pelag- distribution of ocean crust versus transitional crust is ic carbonate deposition ever since. Widespread seismic compatible with a NNW-to-SSE motion of Yucatan out unconformities mark the change from one depositional of the northeastern Gulf, not with a motion from west regime to the next: the mid-Cretaceous unconformity to east. This movement most likely was accommodated correlates with the drowning of carbonate platforms and by some component of counterclockwise rotation of a period of basin starvation, and the late Eocene uncon- Yucatan. formity signals the cessation of Cuban tectonics and sediment input from the south. Numerous smaller hia- CONCLUSIONS tuses and unconformities indicate minor shifts in sedi- 1. A combination of seismic stratigraphy and drill- ment input and dispersal. ing data provides a detailed history of the eastern Gulf 4. Through their stratigraphy and position seaward of Mexico in the Cretaceous and Cenozoic. The data of the Florida-Bahama platform, the basin Sites 535 also present important constraints for the still vaguely and 540 can be correlated with the Camajuani Zone in known earlier stages of the Gulf. central Cuba, whereas the basement Sites 537 and 538 2. Throughout the Cretaceous and Cenozoic, the resemble the basement in the Placetas Zone and its Early southeastern Gulf was a deep seaway between the car- Cretaceous sediment cover. These correlations suggest a bonate platforms of Florida and Yucatan. The structur- deep seaway connecting the southeastern Gulf, northern 733 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR Northwest OBS-12 Oceanic crust 8.1 km/s Figure 16. Line GT2-10C crosses southeastern Gulf from northwest to southeast. Note high-standing basement complex to the south and deeper gra- bens to the north. Basement is overlain by Late Jurassic and Early Cretaceous sediments (drilled in DSDP Holes 535 and 540), indicating that this region of the southeastern Gulf was a deep seaway during Late Jurassic-Early Cretaceous. The line also shows the regional change from ocean crust in the northwest to more faulted and higher-standing transitional crust in the southeast. Northward stratigraphic pinchouts of the inferred Middle and Late Jurassic sequences onto the "outer high" (or ocean crust/transitional crust boundary) suggest a relatively young age (possibly latest Jurassic to earliest Cretaceous) for the ocean crust in the southeastern Gulf. OBS-12 is an ocean-bottom seismometer refraction station (Ibrahim et al., 1981). Cuba, and the central North Atlantic. Atlantic stratig- ACKNOWLEDGMENTS raphy closely resembles that in the southeastern Gulf Much of what we reported could not have been accomplished with- during Berriasian-Barremian time; later, the sites in the out the help of the Scientific Party of Leg 77 and the skilled crews of Gulf were shallower and richer in carbonates than most the Glomar Challenger and the Fred Moore. We thank J. A. Austin and R. Martin for their very helpful and prompt reviews of the manu- of their counterparts in the central North Atlantic. script. This work was supported by the Industrial Associates of the 5. The pre-Cretaceous history of the area must be in- University of Texas Institute of Geophysics and the University of Mi- ferred from seismic stratigraphy and comparison with ami Comparative Sedimentology Laboratory. This is University of the margins of the Gulf of Mexico and the Atlantic. Texas Institute for Geophysics Contribution Number 589. Tilted graben fills on faulted basement may represent REFERENCES Triassic-Early Jurassic rift sediments; up to 2 km of Addy, S. K., and Buffler, R. T., in press. Seismic stratigraphy of the younger Jurassic sediments are interpreted as an upward shelf slope, northeastern Gulf of Mexico. Am. Assoc. Petrol. Geol. succession of alluvial-to-littoral elastics, neritic elastics, Bull. Anderson, T. H., and Schmidt, V. A., 1983. The evolution of Middle and bathyal limestones analogous to the Cayetano, Fran- America and the Gulf of Mexico-Caribbean Sea region during cisco, and Artemisa formations in western Cuba. Mesozoic time. Geol. Soc. Am. Bull., 94:941-966. 6. Basement consists of attenuated and rifted conti- Angstadt, D. M., 1983. Late Cretaceous-Recent seismic stratigraphy nental crust (early Paleozoic metamorphics) dissected by and geologic history of the southeastern Gulf of Mexico—south- Early Jurassic basaltic dikes. western Straits of Florida [M.A. thesis]. University of Texas at Austin. 7. With regard to models on the origin of the Gulf of Angstadt, D. M., Austin, J. A., Jr., and Buffler, R. T, 1983. Deep-sea Mexico, our data set several constraints: There is evi- erosional unconformity in the southeastern Gulf of Mexico. Geol- dence for Triassic-Early Jurassic rifting, but also for ogy, 11:215-218. much later (Late Jurassic-Cretaceous) block faulting, not Arthur, M. A., and Natland, J. H., 1979. Carbonaceous sediments in the North and South Atlantic: the role of salinity in stable stratifi- explained by the models. The sialic crust in the south- cation of Early Cretaceous basins. In Talwani, M., Hay, W. W., eastern Gulf is the same as in Florida, and there is no and Ryan, W. B. F. (Eds.), Deep Drilling Results in the Atlantic support for ocean crust along the postulated Atlantic- Ocean: Continental Margins and Paleoenvironment: Washington Gulf transform link in the area. The offset along the (Am. Geophys. Union), Maurice Ewing Series, 3:375-401. Arthur, M. A., and Schlanger, S. O., 1979. Cretaceous "oceanic anox- postulated transform faults in southern Florida could ic events" as causal factors in development of reef-reservoired gi- not have been more than 100 to 150 km. Yucatan, whose ant oil fields. Am. Assoc. Petrol. Geol. Bull., 63:870-885. motions play a pivotal role in several models, was locked Barnett, R. S., 1975. Basement structure of Florida and its tectonic in place from Early Cretaceous time on. implications. Thins. Gulf Coast Assoc. Geol. Soc., 25:122-142. 734 GEOLOGIC HISTORY - Sites 535/540 Late Jurassic Southeast Early Cretaceous (shallow-deep (deep marine) marine) (EK) (J2) 50 km Figure 16. (Continued). Baudrimont, A. R, and Cussey, R , 1977. Sédimentologie du Juras- Jansa, L. F., Enos, P., Tucholke, B. E., Gradstein, F. M., and Sheri- sique de la Bordure Provençale du Bassin du Sud-est (France). In dan, R. E., 1979. Mesozoic-Cenozoic sedimentary formations of Elf-Aquitaine: Essai de caractérisation sédimentologique des de- the North American Basin; western North Atlantic. In Talwani, pots carbonates. 2. Elements d"interpretation: Pau (Elf-Aquitaine, M., Hay, W. W, and Ryan, W. B. F. (Eds.), Deep Drilling Results Centres de Recherches de Boussens et de Pau), pp. 195-206. in the Atlantic Ocean: Continental Margins and Paleoenviron- Bloxsom, W. E., 1972. A Lower Cretaceous (Comanchean) prograd- ment: Washington (Am. Geophys. Union), Maurice Ewing Series, ing shelf and associated environments of deposition, northern Coa- 3:1-57. huila, Mexico [M.A. thesis]. University of Texas at Austin. Khudoley, K. M., and Meyerhoff, A. A., 1971. Paleogeography and Bryant, W. R., Meyerhoff, A. A., Brown, N. K., Furrer, M. A., Pyle, geological history of greater Antilles. Geol. Soc. Am. Mem., 129: T. E., and Antoine, J. W., 1969. Escarpments, reef trends, and 1-199. diapiric structures, eastern Gulf of Mexico. Am. Assoc. Petrol. Klitgord, K. D., Popenoe, P., and Schouten, H., in press. Florida, a Geol. Bull., 53:2506-2542. Jurassic transform plate boundary. J. Geophys. Res. Buffler, R. T., Shaub, F. J., Huerta, R., Ibrahim, A. B. K., and Lancelot, Y., Winterer, E. L., et al., 1980. Init. Repts. DSDP, 50: Watkins, J. S., 1981. A model for the early evolution of the Gulf Washington (U.S. Govt. Printing Office). of Mexico basin. Oceanol. Acta, Proc. 26th Int. Geol. Cong., Lopez-Ramos, E. L., 1975. Geological summary of the Yucatan Pe- Geol. Continental Margins Symp., 3:129-136. ninsula. In Nairn, A. E. M., and Stehli, F. G. (Eds.), The Ocean Buffler, R. ‰ Watkins, J. S., Worzel, J. L., and Shaub, F. J., 1980. Basins and Margins (Vol. 3): New York (Plenum Press), 257-282. Structure and early geologic history of the deep central Gulf of Malfait, B. T., and Dinkelman, M. G., 1972. Circum-Caribbean tecton- Mexico. In Pilger, R. H. (Ed.), The Origin of the Gulf of Mexico ic and igneous activity and the evolution of the Caribbean Plate. and the Early Opening of the Central North Atlantic Ocean: Baton Geol. Soc. Am. Bull., 83:251-272. Rouge (Louisiana State University), pp. 3-16. Martin, R. G., and Case, J. E., 1975. Geophysical studies in the Gulf Ewing, M., Worzel, J. L., et al., 1969. Init. Repts. DSDP, 1: Washing- of Mexico. In Nairn, A. E. M., and Stehli, F. G. (Eds.), The Gulf ton (U.S. Govt. Printing Office). of Mexico and Caribbean, Ocean Basins and Margins (Vol. 3): Finneran, J. M., Scott, R. W, Taylor, G. A., and Anderson, G. H., New York (Plenum Press), 65-106. 1982. Late Jurassic ramp reefs ("Knowles Limestone"), southwest Pardo, G., 1975. Geology of Cuba. In Nairn, A. E. M., and Stehli, F. flank of the east Texas Basin. In Bebout, D. (Ed.), The Jurassic of G. (Eds.), The Gulf of Mexico and Caribbean, Ocean Basins and the Gulf Rim, Third Ann. Res. Conf., Gulf Coast Sec., Soc. Econ. Margins (Vol. 3): New York (Plenum Press), 553-615. Paleontol. Mineral. Prog, and Abstr., pp. 36-44. Phair, R. L., in press. Lower Cretaceous seismic stratigraphy of the Hall, D. J., Cavanaugh, P. D., Watkins, J. S., and McMillan, K. J., southeastern Gulf of Mexico and southwestern Straits of Florida 1983. The rotational origin of the Gulf of Mexico based on region- [M.A. thesis]. University of Texas at Austin. al gravity data. In Watkins, J. S., and Drake, C. L. (Eds.), Studies Phair, R. L., and Buffler, R. T., 1983. Pre-middle Cretaceous geologic in Continental Margins. Am. Assoc. Petrol. Geol. Mem., 34: history of the deep southeastern Gulf of Mexico. In Bally, A. W. 115-126. (Ed.), Seismic Expressions of Structural Styles—A Picture and Word Heezen, B. C, MacGregor, I. D., et al., 1973. Init. Repts. DSDP, 20: Atlas: Tulsa (Am. Assoc. Petrol. Geol.), AAPG Studies in Geolo- Washington (U.S. Govt. Printing Office). gy, 15(2):2.2.3-141. Hollister, C. D., Ewing, J. I., et al., 1972. Init. Repts. DSDP, 11: Pindell, J., and Dewey, J. F., 1982. Permo-Triassic reconstruction of Washington (U.S. Govt. Printing Office). western Pangea and the evolution of the Gulf of Mexico/Caribbe- Ibrahim, A. B. K., Carye, J., Latham, G., and Buffler, R. T., 1981. an Region. Tectonics, 1:179-211. Crustal structure in the Gulf of Mexico from OBS refraction and Pszczolkowski, A., 1978. Geosynclinal sequences of the Cordillera de multichannel reflection data. Am. Assoc. Petrol. Geol. Bull., 65: Guaniguanico in western Cuba: their lithostratigraphy, facies de- 1207-1229. velopment and paleogeography. Acta Geol. Pol., 28:1-93. 735 W. SCHLAGER, R. T. BUFFLER, D. ANGSTADT, R. PHAIR , 1982. Cretaceous sediments and paleogeography in the west- ern part of the Cuban miogeosyncline. Acta Geol. Pol., 32:135-161. Schlager, W., 1981. The paradox of drowned reefs and carbonate plat- forms. Geol. Soc. Am. Bull., 92:197-211. 538 Schlager, W., Austin, J. A., Jr., Corso, W., McNulty, C. L., Fluegel, Post-Valanginian faulting E., Renz, O., and Steinmetz, J. C, 1984. Early Cretaceous plat- form reentrant and escarpment erosion in the Bahamas. Geology, 12:147-150. Shaub, S. J., Buffler, R. T., and Parsons, J. T, in press. Seismic strati- Platform (Berriasian—Valanginian) graphic framework of the deep Gulf of Mexico basin. Am. Assoc. Petrol. Geol. Bull. Shein, V. S., Ivanov, S. S., Kleshchev, K. A., Khain, V. Y., Marrero, M., and Socorro, R., 1978. Tectonics of Cuba and its shelf. Int. Geol. Rev., 21:540-552. Sheridan, R. E., Crosby, J. T., Bryan, G. M., and Stoffa, P. L., 1981. Stratigraphy and structure of southern Blake Plateau, northern Florida Straits, and northern Bahama Platform from multichan- nel seismic reflection data. Am. Assoc. Petrol. Geol. Bull., 65: 2571-2593. Platform hypothesis Smith, D. L., 1982. Review of the tectonic history of the Florida Base- 537 ment. Tectonophysics, 88:1-22. 538 Sykes, L. R., McCann, W. R., and Kafka, A. L., 1982. Motion of Post-Valanginian faulting Caribbean Plate during last 7 million years and implications for earlier Cenozoic movements. J. Geophys. Res., 87:10656-10676. Thierstein, H. R., 1979. Paleoceanographic implication of organic carbon and carbonate distribution in Mesozoic deep-sea sediment. Aptian—\ Wide platform I In Talwani, M., Hay, W. W., and Ryan, W. B. F. (Eds.), Deep Albian (Berriasian—Valanginian) I Drilling Results in the Atlantic Ocean: Continental Margins and platform Paleoenvironment: Washington (Am. Geophys. Union), Maurice Ewing Series, 3:249-279. Tücholke, B. E., and Mountain, G. S., 1979. Seismic stratigraphy, lith- ostratigraphy and paleosedimentation patterns in the North Amer- ican Basin. In Talwani, M., Hay, W. W., and Ryan, W. B. F. (Eds.), Deep Drilling Results in the Atlantic Ocean: Continental Margins Figure 17. Two contrasting interpretations of Early Cretaceous lime- and Paleoenvironment: Washington (Am. Geophys. Union), Mau- stones at Sites 537 and 538. A. Talus hypothesis assumes that lime- rice Ewing Series, 3:58-85. stones are talus originating at Campeche Bank and shed down a Vail, P. R., Mitchum, R. M., Jr., Todd, R. G., Widmier, J. M., gently dipping seafloor. Talus apron would have to be 30 km wide. Thompson, S., Ill, Sangree, J. B., Bubb, J. N., and Hatlelid, W. B. Platform hypothesis interprets limestones as part of extensive G., 1977. Seismic stratigraphy and global changes of sea level. In Jurassic-Early Cretaceous platform that was partly drowned and Payton, C. E., (Ed.), Seismic Stratigraphy—Applications to Hy- forced to "step back" in Valanginian time. High Aptian-Albian drocarbon Exploration. Am. Assoc. Petrol. Geol. Mem., 26:49-212. platform (present-day Campeche Bank, dotted line) grew on top of Worzel, J. L., Bryant, W., et al., 1973. Init. Repts. DSDP, 10: Wash- this Valanginian platform. Both models call for block faulting be- ington (U.S. Govt. Printing Office). fore and after deposition of the Berriasian-Valanginian limestone interval. Date of Initial Receipt: November 3, 1983 Date of Acceptance: December 2,1983 736 GEOLOGIC HISTORY 538 Position Position of Site 537 of Site 538 Post-Valanginian faulting Pelagic ooze 537 and chalk I_ J j_ J J_ J Pelagic ooze and chalk 1 j_ _i J_ J _ J. Extensive Berriasian-Valanginian platform 1 1 1 1 1 Shallow- Shallow- water water limestone limestone / Shale, \ Berriasian faulting sandstone, i dolomite • /' • /• • Deep-water chalk Alluvial (?) Igneous arkoses o o 5 0 0 basement Basement (phyllite) /fi Figure 18. Stratigraphy of Sites 537 and 538, located on two separate fault blocks 27 km apart. Differences are explained by the interplay of tectonics and sedimentation, shown schematically in center of figure: faulting in Berriasian, followed by undisturbed growth of a carbonate platform (late Berriasian-Valan- ginian) and renewed faulting in post-Valanginian time. 500 km Yaguajay equivalents Carbonate platform Early Cretaceous seaway until Paleogene (mainly deep-water) Carbonate platform Northern limit of until mid-Cretaceous Cuban thrust belt Figure 19. Cretaceous-Tertiary paleogeography of southeastern Gulf of Mexico and postulated relations to facies-structural zones of Cuba. Positions of zones in Cuban thrust belt are schematic and not to scale. 737 -J West East 00 538 CO O 537 0 X mo 1000 T3 jβ 2000 H UI ) D0 € α s 3000 α> •a r E a 3 tt o T 4000 & ö Similarities in basement and lowermost Cretaceous only 3 Jβ •-ö X > PLACETAS CAM A JUAN I CAYO COCO YAGUAJAY Carbonate platform -Maestrichtian sea level- T . r Submarine plateau LK -1000 J × LK -2000 Q. Marginal rise EK ~~ K EK -3000 O U / / / / / / J / / / / / / ~jr~jT-Z~. Deeper-water shale, some Pelagic shale, chert Shallow-water limestones E TZJTLTZI sandstone, some limestone I I Deep-water limestone, chalk, / 7 Shallow-water dolomite | | ooze, some marly limestone with anhydrite Alluvial-littoral sandstone I and conglomerate Figure 20. Correlation of structural-stratigraphic zones in southeastern Gulf of Mexico with facies-structural zones in central Cuba. Schematic cross-sections of Gulf of Mexico based on UTIG seismics and DSDP drilling; stratigraphy of zones in Cuba after Pardo (1975) and Pszczolkowski (1982). This comparison suggests that the eastern margin of the southeastern Gulf and its continuation now exposed on Cuba had very similar topography. Shale and chert in the Placetas Zone of Cuba represent the deepest part of the Cretaceous seaway; no equivalents were recov- ered in the Gulf.